This competing renewal of grant No. P01 CA 33619 "Epidemiologic Studies of Diet and Cancer in Hawaii" continues to explore etiologic factors for cancer and reasons for ethnic variations in cancer incidence. Past cycles of the grant relied heavily, though not exclusively, on case-control studies to explore the relationship of diet to lung, colorectal, endometrial, thyroid, gastric, bladder, skin and other cancers. In addition, we conducted several dietary methodology studies. For this cycle, we propose to take advantage of a recently initiated multi-ethnic cohort in Hawaii to establish a biorepository of blood and urine specimens, and to use this resource to test several hypotheses related to biological markers and selected cancer sites. Over a period of four years, biological samples will be collected from cohort participants of Japanese, Caucasian and Native Hawaiian ethnicity on the island of Oahu, utilizing the widely dispersed facilities of a local clinical laboratory. Based on an anticipated overall response rate of 62 percent, we will obtain samples from 18,535 males and 20,452 females. In addition, a calibration substudy will be conducted to enable us to adjust for measurement error in the biomarkers. The biospecimens will be stored in multiple aliquotso at -150 C (blood) and -80C (urine). Incident cases of colorectal, lung, breast and prostate cancer among the participants will be identified through rapid reporting by the Hawaii Tumor Registry. Blood and urinary markers that will be examined include: 1) vitamin D metabolites, folate and homocysteine, isothiocyanates, insulin and glucose, insulin-like growth factors, and selected genetic markers in relation to colorectal cancer; 2) carotenoids, tocopherols, vitamin C, total phenol, isothiocyanates, selenium, and isoprostane in relation to lung cancer; 3) phytoestrogens, carotenoids, tocopoherols, insulin-like growth factors, estrogens, and androgens in relation to breast cancer; and 4) selenium, phytoestrogens, carotenoids, tocopoherols, fatty acids, insulin-like growth factors, isoprostane, and androgens in relation to prostate cancer. All four projects will use a nested case-control design and will be supported by four Core Components: 1) Administration, 2) Data Collection, 3) Biospecimens, and 4) Data Management and Analysis. By taking advantage of a unique resource in Hawaii, namely, the Multi-ethnic Cohort Study, this project will help address certain timely issues related to the nutritional etiology of cancer.